ADOBSVM: Anomaly detection on block chain using support vector machine

Abstract

Bitcoins are cryptocurrencies that make use of blockchain technology, which consists of network nodes and permanent ledgers of events. Theft and illicit actions are referred to as abnormalities in the financial network. Members of the network want to figure out what's up as quickly as possible to avoid conflicts in the community and maintain the network's integrity. Using an unconstrained machine learning technique, this research aims to discover anomalies in bitcoin transactions. Anomaly detectors play a critical role in defending networks and systems from unexpected threats, typically by identifying as well as screening problems efficiently. Various strategies have been developed over the years, all with the goal of lowering the number of false positives. Furthermore, no proposals so far talked about blockchain-based system-based attacks were considered. The Support Vector Machine (SVM) for anomalies is described in this article. In comparison to other existing approaches, SVM provides good security with a shorter execution time. The effectiveness of the SVM approach is evaluated using attack detection rate, error rate, execution time, and power consumption; and it is compared to existing methods. According to experimental results, the best results from existing approaches are obtained using SVM.

Keywords

• Blockchain
• Anomaly detection
• Machine learning
• SVM
• KNN
• NB